AN APPROACH FOR COLOR IMAGE COMPRESSION OF BMP AND TIFF IMAGES USING DCT AND DWT

Transcription

1 INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING & TECHNOLOGY (IJCET) International Journal of Computer Engineering and Technology (IJCET), ISSN (Print), ISSN (Print) ISSN (Online) Volume 6, Issue 1, January (2015), pp IAEME: Journal Impact Factor (2014): (Calculated by GISI) IJCET I A E M E AN APPROACH FOR COLOR IMAGE COMPRESSION OF BMP AND TIFF IMAGES USING DCT AND DWT A.H.M. Jaffar Iqbal Barbhuiya 1, Tahera Akhtar Laskar 2, K. Hemachandran 3 1, 2, 3 Department of Computer Science, Assam University, Silchar, Assam, India ABSTRACT Now a days image compression has become is an indispensible part of digitized image storage and transmission. Compression of an image is necessary before storing and transmitting it due to its limitation of storage and bandwidth capacity. Wavelet transform decomposes complexion of images into its elementary forms. In this paper, a comparative study has been carried out on image compression using DCT (Discrete Cosine Transform) and DWT (Discrete Wavelet Transform). A comparison is outlined to emphasize the results of this compression system between DCT and DWT using BMP (Bitmap ) and TIFF (Tag Index File Format) color images. We have done conversion of color images into gray scale and also compression of gray scale image is shown after conversion using DWT method. DWT algorithm performs much better than DCT algorithms in terms of Mean Square Error (MSE) and Peak Signal to Noise Ratio (PNSR). Keywords: DCT; DWT ; PSNR; MSE; Wavelet Transform; Compression; Huffman Coding; BMP; TIFF; Compression ratio. I. INTRODUCTION An image can be defined as a matrix of pixel or intensity values. compression is used to reduce the redundancy and randomness present in the image because to increase the storing capacity and efficiency level of the images. Therefore it is essential to compress the images by storing only the required information needed to reconstruct the image. To compress any image, redundancy must be removed. Sometimes images having large areas of same color will have large redundancies and similarly images that have frequent and large changes in color will be less redundant and harder to compress[1-3]. 19

2 The main objective of this paper is to reduce irrelevance and redundancy of the JPEG and PNG image data in order to be able to store or transmit data in an efficient form using DCT and DWT. We have tried to study the different image compression algorithm and evaluate their performance on different image formats and also developed a system for image compression using DWT and compare the results with the existing techniques or systems. compression can be done in two ways: Lossy Compression Lossless Compression If any pixel value is changed from a digital image and then energy will be lost and this technique is called lossy compression. The amount of information retained by an image after compression and decompression is known as lossless compression [7]. In this research paper, a modified and competent image compression scheme is proposed based on DWT and which results a good compression ratio without degrading the quality of the image. The proposed algorithm has been analyzed and compared with some other existing method. Several quality measurement techniques like PSNR and MSE have been considered to determine the image compression with respect to the reference image. For that purpose some traditional images are taken and have done experiment on it [1-3]. This paper structure contains the following sections: section 2 describes the previous work done by different authors, section 3 describes the basic principles of image compression techniques and related works, section 4 illustrates the fundamentals of wavelet transform, section 5 presents the proposed method using DWT, section 6 demonstrates the experimental results and discussions and finally section 7 concludes the paper. II. RELATED WORKS compression has many practical applications because of huge data storage, transmission and retrieval for medical imaging, documents and videoconferencing. Chowdhury and Khatun [1] introduced the performance of the image compression algorithm and provide sufficient high compression ratios compared to other techniques. Sukanya and Preethi [2] has explained about compression methods such as JPEG 2000,EZW,SPHIT (Set Partition in Hierarchical Trees) and HS- SPHIT (Highly Scalable-SPHIT) on the basis of hierarchical time,error comparison, MSE, PSNR and compression ratio. Qureshi [9] in his Computational Project image compression using wavelet transform described how wavelets can be used in image compression. Lawson and Zhu [12] introduced DWT as a tutorial form and illustrates how it how it may be used as a part of of an image coding algorithm. Averbuch et al. [3] has tested several compression techniques and 512X512 images are trained together and common table codes were created to achieve a compression ratio and a PSNR of and compression ratio and a PSNR of Talukdar and Harada [4] in their paper explained the 2D Discrete Wavelet Transform (DWT) has been applied and the detail matrices from the information matrix of the image have been estimated. Wavelet transform is used in the reconstructed image and is amalgamated by the estimated detail matrices and information matrix. The compression quality of image has been calculated using Compression Ratio (CR), Peak Signal to Noise Ratio (PSNR), Mean Opinion Score (MOS), and Picture Quality Scale (PQS). III.IMAGE COMPRESSION TECHNIQUES The basic encoding method for transform based compression works as follows: 20

3 1) transform: Divide the source image into blocks and apply the transformations to the blocks. 2) Parameter quantization: The data generated by the transformation are quantized to reduce the amount of information. This step represents the information within the new domain by reducing the amount of data. Quantization is in most cases not a reversible operation because of its lossy property. 3) Encoding: Encode the results of the quantization. This last step can be error free by using Run Length encoding or Huffman coding. It can also be lossy if it optimizes the representation of the information to further reduce the bit rate. Transform based compression is one of the most useful applications. Combined with other compression techniques, this technique allows the efficient transmission, storage, and display of images that otherwise would be impractical [4]. 4) DCT-Based Transform Coding: The DCT was first applied to image compression in the work by Ahmed et al [5]. DCT is a transform coding method comprising four steps. The source image is first partitioned into sub-blocks of size 8x8 pixels in dimension. Then each block is transformed from spatial domain to frequency domain using a 2-D DCT basis function. The resulting frequency coefficients are quantized and finally output to a lossless entropy coder. DCT is an efficient image compression method since it can decorrelates pixels in the image and compact most image energy to a few transformed coefficients. BMP and TIFF may be replaced by wavelet-based image compression algorithms, which have better compression performance [5]. IV. THE WAVELET TRANSFORM A wavelet is a mathematical function used to divide a given function or continuous-time signal into different scale components. It can assign a frequency range to each scale component. All scale components that matches its scale can be studied with a resolution. It is the delegation of a function by wavelets., =,,, 0, Where ψ is a wavelet function, a, is another function which measure the degree of compression, and b, is a translation function which measures the time location of the wavelet [8]. A. Wavelet Definition: Wavelets are mathematical functions that cut up data into different frequency components. The fundamental idea behind the wavelets is to analyze the signal into different scales and resolutions [2]. B. Quantization: Quantization is the process where actual reduction of image is done. It is a lossy compression technique which basically used in DCT data quantization and DWT data quantization for BMP and TIFF images. C. Entropy Encoding: Entropy encoding is a lossless data compression method. Two commonly used entropy encoding techniques are Huffman coding and Arithmetic coding. D. Discrete Wavelet Transform (DWT): Suppose a function f(x) is a continuous function and is expanded in a sequence of numbers, then the resulting coefficients are called Discrete Wavelet Transform (DWT) of f(x). 21

4 LL LH HL HH 1 st Level LL HL HL LH HH LH HH 2 nd Level LL HL HL HL LH HH LH HH LH HH 3 rd Level Figure 1. Wavelet Decomposition The discrete wavelet transform in two dimensions of functions f(x,y) of size M N is [6] W φ j,m,n = 1 MN f x,y φ,, x,y The DWT can be decomposed into three level of decomposition. For example, decomposition level 1, decomposition level 2 and decomposition level 3. The above Fig. 1 shows each level of decomposition. Compression ratio of an image can be calculated using the following formula: Compression Ratio (%) =Original image size-compressed image size/original image*100 V. PROPOSED COMPRESSION METHOD USING DWT This portion describes the compression method based on DWT. The basic method of the wavelet transform is selecting a color input image of any format from the computer and to find out the size of the image, compressed size of the image, percentage of compression using DCT and percentage of compression using DWT and then convert the compressed image into gray scale and finally find the compression ratio of gray scale image and also find out the PSNR and MSE of the image using DWT and IDWT of three level of decomposition. The method is given below: Read an input image of any format from the user. We are using 2D wavelet decomposition with respect to a haar or doubencies wavelet computes the approximation coefficients matrix and detail coefficient matrixes horizontal, vertical & diagonal respectively which is obtained by wavelet decomposition of the input matrix. Again using 2D wavelet decomposition with respect to a haar or doubencies wavelet computes the approximation and detail coefficients which are obtained by wavelet decomposition of the approximation coefficient matrix. This is considered as level 2. Again if we apply the haar wavelet transform from approximation coefficient matrix which is considered as approximation coefficient 1 for level 3. Again we will take inverse transform for level 1, level 2 & level 3 that is, image input and approximation coefficients respectively for each level of decomposition. We can reconstruct the images for level 1, level 2 & level 3. After this whole process it will display the results of reconstruction 1, reconstruction 2 and reconstruction 3 that is, level 1, 2 and 3 with respect to the original image. The Fig. 2 shows the basic compression method of our work. 22

5 Wavelet Transform Quantization Entropy Decoding Compressed Entropy Encoding Dequantization Inverse Wavelet Transform Reconstructed Figure 2. The structure of image compression using wavelet transforms VI. RESULTS AND DISCUSSIONS In this research, an efficient compression method based on DWT is proposed. The algorithm has been implemented in MATLAB A set of test images like BMP and TIFF are taken to justify the efficiency of the algorithm. Fig 3 shows the test images and their respective resulting compressed images using BMP, TIFF and the proposed compression methods. The experimental results with the proposed compression method have been arranged in the Table 1 and Table 2 for the different image processing images. It also shows the comparison between DCT and the proposed compression method. From these tables, we find out the compression ratio of different BMP and TIFF color and gray scale images with different resolutions and then finally to find the PSNR and MSE value of images using DWT. So, the proposed compression technique gives the better results compared to other compression methods. a)original barbara image(bmp) b)compressed image (DWT) c) Compressed image (DCT) d) Compressed gray scale image Resolution: 512*512 Figure 3. A barbara image and corresponding compressed images in BMP format using DCT and DWT methods. 23

6 a)original peppers image(tiff) b)compressed image(dwt) c) Compressed image (DCT) d) Compressed gray scale image Resolution: 512*512 Figure 4. A pepper image and corresponding compressed images in TIFF format using DCT and DWT methods. Figure 5: Barbara image at decomposition Figure 6: Peppers image at decomposition level 1 level 1 24

7 Original Table I: Experimental Results of BMP s at Decomposition Level 1 Resolution Deco mpo sitio n Leve l Original Size (bytes) Compressed Color size Propose d DCT Method (DWT) Compressi on Ratio of Color image (%) Compre ssed Graysca le image Size Compressi on Ratio of Grayscale image(%) Zelda256.bmp 256X Zebra256.bmp 256X House.bmp 256X Baboon256.bmp 256X Barbara512.bmp 512X Peppers.bmp 512X Lenna128.bmp 128X airplane.bmp 512X Boat.bmp 512X sailboat.bmp 512X MSE (%) PSNR (db) Original Table II: Experimental Results of TIFF s at Decomposition Level 1 Deco Original Compressed Color Compres Compres mpo Size size (bytes) sion sed sitio (in bytes) DCT Propose Ratio of Grayscal n d Color e image Leve Method image Size(byte l (DWT) (%) s) Resolution Compres sion Ratio of Grayscal e image (%) Zelda256.tiff 256X Zebra256.tiff 256X House.tiff 256X Baboon64.tiff 64X Barbara512.tiff 512X Peppers.tiff 512X lena_512.tiff 512X RGB.tiff 256X Woman_baby.tiff 512X strawberries.tiff 666X MSE (%) PSNR (db) VII. CONCLUSION In this research plan, an attempt has been made to study and compare the image compression techniques using DCT and DWT. From the above experimental analysis, it is seen that in comparison to the other experiments our experiment shows better performance with respect of compression size and percentage of image compression. A new algorithm has been proposed on Compression using DWT and Inverse DWT. An experimental result has been shown after compressing any color image of different image formats. The most distinguishing feature of using DWT and Inverse DWT is that it will not only enable to compress an image but also will help to maintain the quality of the image as it was in its original form, which was not possible earlier in other image compression techniques. The future direction of this research is to develop an algorithm using DCT or DWT where any random image of any resolution or size could be compressed at a uniform rate without degrading the quality of image. 25

8 REFERENCES 1. M.M.Chowdhury,and A.Khatun, Compression using Discrete Wavelet Transform, International Journal of Computer Science Issues, Vol.9, Issue4, No.1, July 2012, ISSN (online): , 2. Y.Sukanya, and J.Preethi, Analysis of Compression algorithm using Wavelet Transform with GUI in MATLAB, IJERT: International Journal of Research in Engineering and Technology, eissn: , pissn: , volume: 02, Issue: 10, Oct-2013, , A.Averbuch,D.Lazar, and M.Israeli, Compressi on using Wavelet Transform and Multiresolution decomposition,ieee Transactions on Processing, Vol.5,No.1, January, K.H.Talukder, and K. Harada, Haar Wavelet Based Approach for Compression and Quality Assessment of Compressed, IAENG International Journal of Applied Mathematics, 36:1, IJAM_36_1_9, Advance Online Publication: 1 February, N.Ahmed,T.Natarajan, and K.R. Rao, Discrete Cosine Transform,IEEE Transactions on Computers, January, 1974, pp R.C.Gonzalez,R.E.Woods,Digital Processing, 2 nd Edition, Chapter 7, Wavelet and Multiresolution Processing, pp B.Nilesh,S.Sachin,N.Pradip, and D.B.Rane, Compression using DWT,ISSN: , IJCTEE,Volume3,Special Issue,March-April A.H.M.J.I.Barbhuiya,M.Singha, and K.Hemachandran, Wavelet Transformation and Its Applications in Digital Processing, In the proceeding of National Conference on Latest Trends and Innovations in Computing, Electronics & Communication, held at AISECT University, Bhopal, November, 17-18, F.Z. Qureshi," Compression Using Wavelet Transform", Computational Vision Project, University of Toronto, S.Anandanarayanan, and S.K.Srivatsa, A High Performance Novel Compression Technique Using Huffman Coding with Edge Dection, International Journal of Computer Engineering & Technology (IJCET), ISSN (Print), ISSN (Online), Volume 4, Issue 2, March April (2013), pp , IAEME: A.H.M.J.I.Barbhuiya, and K.Hemachandran, Wavelet Transformations and Its Major Applications in Digital Processing, International Journal of Engineering Research & Technology (IJERT), ISSN: , Vol. 2 Issue 3, March, S.Lawson and J.Zhu, Compression using wavelets and JPEG2000: a tutorial, Electronics and Communication Engineering Journal, June Laskar, Tahera Akhtar, and K. Hemachandran. "Digital Watermarking Techniques and Its Applications." International Journal of Engineering Research and Technology. Vol. 2. No. 3 (March-2013). ESRSA Publications, Pardeep Singh, Nivedita, Sugandha Sharma, A Comparative Study: Block Truncation Coding, Wavelet, Embedded Zerotree and Fractal Compression on Color International journal of Computer Engineering & Technology (IJCET), Volume 3, Issue 2, 2012, pp , ISSN Print: , ISSN Online: Hitashi and Sugandha Sharma, Fractal Compression Scheme Using Biogeography Based Optimization on Color s International journal of Computer Engineering & Technology (IJCET), Volume 3, Issue 2, 2012, pp , ISSN Print: , ISSN Online: